- •1 Purpose, structure and classification of error-control codes
- •1.1 Error-control codes in transmission systems
- •1.2 Classification of error-control codes
- •Questions
- •2 Parameters of block error-control codes
- •Questions
- •3 Error detection and correction capability of block codes
- •4 Algebraic description of block codes
- •5.2 Syndrome decoding of the block codes
- •5.3 Majority decoding of block codes
- •Questions
- •6 Boundaries of block codes parameters
- •6.1 Hamming upper bound
- •6.2 Varshamov-Gilbert lower bound
- •6.3 Complexity of coding and decoding algorithms
- •Questions
- •7 Important classes of block codes
- •7.1 Hamming codes
- •7.2 Cyclic codes
- •Questions
- •8 Decoding noise immunity of block codes
- •8.1 Decoding noise immunity of block codes
- •8.2 Energy coding gain
- •Questions
- •9 Structure and characteristics of convolutional codes
- •9.1 Description methods of convolutional codes
- •9.2 Key parameters and classification of convolutional codes
- •Questions
- •10 Decoding algorithms of convolutional codes
- •10.1 Classification of decoding algorithms
- •10.2 Viterbi algorithm for decoding of convolutional codes
- •Questions
- •11 Noise immunity of convolutional code decoding
- •11.1 Decoding error probability of convolutional code
- •11.2 Energy coding gain
- •12.2 Limiting efficiency of transmission systems and Shannon bound
- •12.3 Perspective ways of further increasing efficiency
- •Attachment а. Performances of error-correcting codes а.1 Performances and generator polynomials of cyclic codes
- •А.2 Energy coding gain by using of the cyclic codes
- •А.3 Performances of binary convolution codes
- •Attachment b. Methodical manual for the course work
- •It is necessary:
- •Methodical instructions
- •Example of calculations and code optimisation procedure
- •Input data:
- •3 Questions
- •4 Home task
- •5 Laboratory task
- •6 Description of laboratory model
- •Questions
- •4 Home task
- •5 Laboratory task
- •6 Description of laboratory model
- •7 Requirements to the report
- •Lw 4.3 Noise immunity of block error-control codes researching
- •1 Objectives
- •2 Main positions
- •2.3 Coding gain
- •3 Questions
- •4 Home task
- •5 Laboratory task
- •6 Description of the computer program of (n, k) code correcting ability research
- •7 Requirements to the report
- •Lw 4.4 Studying of coding and decoding by error-control convolution codes
- •1 Objectives
- •2 Main principles
- •3 Questions
- •4 Home task
- •5 Laboratory task
- •6 Description of laboratory model
- •7 Requirements to the report
- •Attachment d. Dictionaries d.1 English-Russian dictionary
- •D.2 Russian-English dictionary
- •References
- •Ivaschenko Peter Vasilyevich
- •Bases of the error-control codes theory Education manual
12.2 Limiting efficiency of transmission systems and Shannon bound
Indicators and make sense a specific rates, and inverse values =1/ and =1/ define specific expenses of corresponding resources on an information transferring with unity rate (1 bit per second). For the Gaussian channel with frequency band Fch, the ratio of signal to noise =Ps/Pn and channel capacity it is possible to establish, that these efficiency indicators are connected by the relation:
and (12.4)
For ideal system (=1) limiting equation can be defined. According to Shannon theorem by the corresponding transmission methods (coding and modulation) and receiving (demodulation and decoding), the value can be as much as close to unit. Thus the error can be made as much as small. In this case by a condition = 1 it is received limiting equation between and :
. (12.5)
This formula defines of energy efficiency from the frequency efficiency for the ideal system ensuring equality of a information rate to a channel capacity. It is convenient to represent this equation in the form of a curve on a plane = f() (figure 12.1, a curve «Shannon bound»). This curve is limiting and reflects the best interchanging between and in the continuous channel (СC).
It is necessary to notice, that frequency efficiency varies in limits from 0 to , energy efficiency is bounded above by magnitude:
. (12.6)
Differently, energy efficiency of any information transmitting system in a Gaussian channel can not exceed the magnitude
. (12.7)
Similar limiting curves can be constructed and for any other channels if in formulas (12.2) and (12.3) instead of a rate Rchan to substitute expressions for a channel capacity of the corresponding channel. So, in particular, on fig. 12.1 the curve for limiting equation =f () the is discrete-continuous channel (D-CC) is shown.
It "is enclosed" in a curve of the continuous channel (CC) that confirms know result of an information theory according to which DN channel capacity of D-CC always is less a than channel capacity of the continuous channel (CC) which is a basis for construction of corresponding D-CC. In real digital systems error probability p always has a final value and informational efficiency is less then a limiting value max. In these cases for the fixed error probability p = const it is possible to define efficiency ratio to and to construct curves = f ().
In coordinates (, ) to each variant of a transmission system there will corresponds a point on a plane. All these points (curves) should place below a limiting curve of «Shannon bound». The place of these curves depends on an aspect of signals (modulations), a codes (coding methods) and a method of the elaborating of a signals (demodulation/decoding). About perfection of the digital telecommunication methods judge on a degree of placing of real efficiency of to the limiting values.
Concrete data about the efficiency of various modulation/coding methods and also their combinations are given in following section.